make word recount a tokenizer-specific function

[nominatim.git] / nominatim / clicmd / replication.py

diff --git a/nominatim/clicmd/replication.py b/nominatim/clicmd/replication.py
index f9c5561a5c53edd099bb0855e9a58562335e3514..a22cef478d7459edbbef7a7cb5536d016e1948f5 100644 (file)
--- a/nominatim/clicmd/replication.py
+++ b/nominatim/clicmd/replication.py
@@ -6,9 +6,9 @@ import logging
 import socket
 import time
 
-from ..db import status
-from ..db.connection import connect
-from ..errors import UsageError
+from nominatim.db import status
+from nominatim.db.connection import connect
+from nominatim.errors import UsageError
 
 LOG = logging.getLogger()
 
@@ -20,6 +20,19 @@ LOG = logging.getLogger()
 class UpdateReplication:
     """\
     Update the database using an online replication service.
+
+    An OSM replication service is an online service that provides regular
+    updates (OSM diff files) for the planet or update they provide. The OSMF
+    provides the primary replication service for the full planet at
+    https://planet.osm.org/replication/ but there are other providers of
+    extracts of OSM data who provide such a service as well.
+
+    This sub-command allows to set up such a replication service and download
+    and import updates at regular intervals. You need to call '--init' once to
+    set up the process or whenever you change the replication configuration
+    parameters. Without any arguments, the sub-command will go into a loop and
+    continuously apply updates as they become available. Giving `--once` just
+    downloads and imports the next batch of updates.
     """
 
     @staticmethod
@@ -38,13 +51,13 @@ class UpdateReplication:
                            help=("Download and apply updates only once. When "
                                  "not set, updates are continuously applied"))
         group.add_argument('--no-index', action='store_false', dest='do_index',
-                           help=("Do not index the new data. Only applicable "
+                           help=("Do not index the new data. Only usable "
                                  "together with --once"))
         group.add_argument('--osm2pgsql-cache', metavar='SIZE', type=int,
                            help='Size of cache to be used by osm2pgsql (in MB)')
         group = parser.add_argument_group('Download parameters')
         group.add_argument('--socket-timeout', dest='socket_timeout', type=int, default=60,
-                           help='Set timeout for file downloads.')
+                           help='Set timeout for file downloads')
 
     @staticmethod
     def _init_replication(args):
@@ -55,8 +68,7 @@ class UpdateReplication:
             replication.init_replication(conn, base_url=args.config.REPLICATION_URL)
             if args.update_functions:
                 LOG.warning("Create functions")
-                refresh.create_functions(conn, args.config, args.sqllib_dir,
-                                         True, False)
+                refresh.create_functions(conn, args.config, True, False)
         return 0
 
 
@@ -84,6 +96,7 @@ class UpdateReplication:
     def _update(args):
         from ..tools import replication
         from ..indexer.indexer import Indexer
+        from ..tokenizer import factory as tokenizer_factory
 
         params = args.osm2pgsql_options(default_cache=2000, default_threads=1)
         params.update(base_url=args.config.REPLICATION_URL,
@@ -93,7 +106,7 @@ class UpdateReplication:
                       indexed_only=not args.once)
 
         # Sanity check to not overwhelm the Geofabrik servers.
-        if 'download.geofabrik.de'in params['base_url']\
+        if 'download.geofabrik.de' in params['base_url']\
            and params['update_interval'] < 86400:
             LOG.fatal("Update interval too low for download.geofabrik.de.\n"
                       "Please check install documentation "
@@ -107,6 +120,8 @@ class UpdateReplication:
                 raise UsageError("Bad argument '--no-index'.")
             recheck_interval = args.config.get_int('REPLICATION_RECHECK_INTERVAL')
 
+        tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config)
+
         while True:
             with connect(args.config.get_libpq_dsn()) as conn:
                 start = dt.datetime.now(dt.timezone.utc)
@@ -114,10 +129,11 @@ class UpdateReplication:
                 if state is not replication.UpdateState.NO_CHANGES:
                     status.log_status(conn, start, 'import')
                 batchdate, _, _ = status.get_status(conn)
+                conn.commit()
 
             if state is not replication.UpdateState.NO_CHANGES and args.do_index:
                 index_start = dt.datetime.now(dt.timezone.utc)
-                indexer = Indexer(args.config.get_libpq_dsn(),
+                indexer = Indexer(args.config.get_libpq_dsn(), tokenizer,
                                   args.threads or 1)
                 indexer.index_boundaries(0, 30)
                 indexer.index_by_rank(0, 30)
@@ -125,6 +141,7 @@ class UpdateReplication:
                 with connect(args.config.get_libpq_dsn()) as conn:
                     status.set_indexed(conn, True)
                     status.log_status(conn, index_start, 'index')
+                    conn.commit()
             else:
                 index_start = None